Every few months, another “how to set up Obsidian so it really works for you” essay does the rounds.
This morning I found yet another one. It looked good. Genuinely thorough. The kind of post where someone has clearly spent weeks refining their setup and another week writing about it.
And I had absolutely no idea what to do with it.
Do I read it now? File it somewhere? But where? The irony of not knowing how to file a productivity system article inside your productivity system is not lost on me.
Anyway. I just threw it at Claude with a simple prompt: review this, tell me what we’re already doing, and flag anything worth adding.
It came back having identified that roughly 80% of the article described things already baked into our setup. Which is both validating and a little amusing - apparently we’ve been doing this right without knowing it was fashionable. For the remaining 20%, it pulled out a handful of ideas that would genuinely extend what we’ve already got.
I asked for an implementation plan.
Claude Code came back with something along the lines of: here’s what I’d do - create a few slightly different structures, move some files around, write up a handful of new documents. Clean, concrete, sequenced.
I said: excellent, go ahead and document it as part of our Personal OS.
Done.
Here’s the thing I figured out about myself through all of this.
I am very good at systems thinking. Give me a problem and I’ll have a framework for it before you’ve finished the sentence. I find this genuinely enjoyable.
The doing is where it falls apart.
The prospect of actually going into a filing system and tidying it up has roughly the same appeal as alphabetising my bookshelf or reorganising my sock drawer. I want the outcome. I have zero interest in the process of getting there. The tedium of it, the clicking, the renaming, the deciding where things go - it’s the kind of work that makes me want to do something else immediately.
Fortunately, Claude Cowork has an apparently endless appetite for exactly this kind of work.
The pairing turns out to be fairly natural. I do the thinking, Claude does the doing. I identify that something should change, Claude figures out the steps and executes them. Neither of us is doing the other’s job. We’re just covering each other’s gaps.
I suspect I’m not the only person built this way. Systems thinking is a reasonably common trait. Systems doing - the patience to actually implement, maintain and iterate on those systems - is much rarer. Most productivity advice assumes you have both. Most people, if they’re being honest, have one or the other.
The gap between what you know should happen and what you can be bothered to make happen is where most good ideas go to die.
It’s considerably smaller now than it used to be.
On Obsidian: If you’re not already using it, Obsidian is open source and free for Mac and Windows. It’s also the easiest way I’ve found to set up a file system that works naturally with Claude CoWork. You get search, structure and a proper folder hierarchy - but underneath all of that, the files are just markdown. Which means if Obsidian ever disappears, your data doesn’t go with it.
For those who want to go deeper, Nick Milo’s Linking Your Thinking is worth bookmarking. He’s one of the more accomplished educators in this space and has built a substantial body of work around making Obsidian genuinely useful rather than just clever.
That said, as I’ve demonstrated above, you don’t need to get particularly sophisticated. CoWork can build and maintain the structures for you. The system thinking is still yours. The system doing, increasingly, isn’t something you have to worry about.
PS. A link to the original article that prompted this